INDEX

Explanations

all samples, patients, groups, teams, studies

np_acts-logits-general · gemini-2.5-flash-lite

patterns where "patients" or "groups" appear in scientific or medical contexts, often followed by verbs like "were" or "had".

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron detects mentions of study‐participant or sample counts—i.e. nouns like “samples,” “patients,” “groups,” “teams,” or “methods” when they refer to numbered items in a scientific context.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

not

-1.42

 ۱۶

-1.19

 faſt

-1.18

achelors

-1.17

 brug

-1.16

험

-1.13

melons

-1.13



-1.13

 mors

-1.11

 CACHE

-1.09

POSITIVE LOGITS

 individuales

1.42

飲

1.34

 έχουν

1.33

 INCLUDING

1.26

Алек

1.23

 individual

1.23

ive

1.19

みんなの

1.17

日々

1.17

 since

1.16

Activations Density 0.108%

all samples, patients, groups, teams, studies

patterns where "patients" or "groups" appear in scientific or medical contexts, often followed by verbs like "were" or "had".

The neuron detects mentions of study‐participant or sample counts—i.e. nouns like “samples,” “patients,” “groups,” “teams,” or “methods” when they refer to numbered items in a scientific context.

No Comments

No Known Activations

all samples, patients, groups, teams, studies

patterns where "patients" or "groups" appear in scientific or medical contexts, often followed by verbs like "were" or "had".

The neuron detects mentions of study‐participant or sample counts—i.e. nouns like “samples,” “patients,” “groups,” “teams,” or “methods” when they refer to numbered items in a scientific context.

No Comments

No Known Activations